3.1.3. Antioxidant Activity

Three methods were applied to determine the antioxidant activities in the EOP extracts. The experimental values (Table 2) varied between 40.5 and 59.9 mg TE/g EOP in the DPPH assay, between 72.1 and 140.5 mg TE/g EOP in the ABTS assay, and between 36.1 and 64.4 mg TE/g EOP in the FRAP assay. The software generated similar model equations for DPPH Equation (4), ABTS Equation (5), and FRAP assays Equation (6).

In the three analytical methods that were applied, the antioxidant activity depended on the three variables studied, with the linear and quadratic terms for the ethanol concentration being the most significant ones with a clear negative influence. It is worth noting that a significant interaction e ffect between this factor and the ultrasound amplitude was detected for ABTS and FRAP assays, but not for DPPH. In the case of ABTS assay, the negative coe fficient for the quadratic term for extraction time was not su fficient to change the trend. Figure 4a,c,e depict the influence of ethanol concentration and amplitude on the DPPH, ABTS, and FRAP assay responses, respectively. Figure 4b,d,f present the surface responses as a function of amplitude and extraction time for the same responses. The three assays of antioxidant activity produced similar influences of the studied variables on the TPC and TFC

responses. This behaviour indicates a clear correlation between the presence of phenolic compounds and the antioxidant activity of the EOP extracts. The extraction time and amplitude increased with the values of the responses across the entire studied ranges, and the ethanol concentration produced maximum yield when the liquid was approximately 40% ethanol.

**Figure 4.** Response surfaces of the antioxidant activity of the extracted olive pomace (EOP) extracts. (**<sup>a</sup>**,**b**) DPPH assay; (**<sup>c</sup>**,**d**) ABTS assay; and, (**<sup>e</sup>**,**f**) FRAP assay.

### *3.2. Process Optimisation and Validation of the Model*

An optimisation of the three studied variables was carried out with the objective of simultaneously maximising the five measured responses due to the relationship between phenolic compounds and their antioxidant activity (i.e., TPC, TFC, DPPH, ABTS, and FRAP). The optimal conditions that were predicted by the model that maximised all responses included a 43.2% ethanol concentration, 70% amplitude, and a 15 min extraction time. Table 4 shows the values that were predicted by the model for all of the responses in the optimal conditions and the means of three experiments that were performed in the optimal conditions to test the adequacy of the model. The experimental data were close to the values that were predicted by the model; the error was below 10% in all cases.

**Table 4.** Predicted and experimental ultrasound-assisted etraction (UAE) values of EOP in the optimal conditions that simultaneously maximised the five responses.


Regarding other residues from olive oil production, Martinez-Patiño et al. [17] studied the phenolic content and antioxidant activity of extracts from olive tree pruning (OTP) and olive mill leaves (OML) while also using UAE. The optimised UAE conditions for OTP and OML were similar to those for EOP, but the ethanol concentration was slightly higher (approximately 55%). All of the responses that were determined for the EOP extracts were significantly greater than those of the OTP and OML extracts. Table 5 summarises the increments in the five studied responses in the EOP extracts with respect to OTP and OML. In the case of TPC, the EOP liquids produced results that were 1.85 times greater than those from the OTP extracts and 1.37 times greater than those of the OML extracts.

**Table 5.** Comparison of the main antioxidant activity indicators for EOP and other olive oil production residues.


Other by-products from agro-industrial processes have been studied in terms of obtaining phenolic compounds while also using UAE. For example, waste from sunflower oil production, e.g., sunflower seed cake, reached 17.96 mg GAE/g dry TPC biomass in optimised UAE conditions, 43% ethanol, 70 ◦C and 86 μm amplitude [20], and the residue from lemon juice production produced 17.97 mg GAE/g dry TPC biomass and an antioxidant activity of 9.4 mg TE/g dry biomass in a FRAP assay of extracts that were obtained with UAE at 50 ◦C, 45 min and 250 W [21]. In both cases, the reported phenolic compound concentrations are lower than those that were obtained with EOP extracts in this work. Hence, EOP could be considered to be a relevant source for antioxidant compounds, in an olive-derived biorefinery concept.

### *3.3. Olive Pomace and Extracted Olive Pomace as Sources of Antioxidants*

Olive pomace (OP) and extracted olive pomace (EOP), respectively, are the main wastes in the olive mills and the olive pomace extracting industries. Table 6 presents recent references from the literature related to the extraction of phenolic compounds from these agro-industrial residues. OP is the by-product obtained after olive oil separation (by centrifugation), which still contains residual oil and it is described with different names that include alperujo or olive cake while orujillo is the final residue, exhausted, and dry olive pomace. Most of these references used olive pomace from olive mills as raw material. In most of these works, olive pomace from the olive mills was used as raw material, and then it was defatted in the laboratory with an organic solvent prior to antioxidant extraction. Only the present work and that of Caballero et al. [22] used real EOP industrial waste.


**Table 6.** Antioxidant capacity indicators for olive pomace extracts.

TPC: total phenolic content; GAE: gallic acid equivalents; d.w.: dry weight; TE: trolox equivalents; TFC: total flavonoids content; HT: hydroxytyrosol; MA: maslinic acid; OA: oleanolic acid; EE: epicatechin equivalents; FSE: ferrous sulfate equivalents; CTE: catequin equivalents.

Several extraction methods have been studied, including conventional extraction, UAE, microwaveassisted extraction (MAE), high hydrostatic pressure-assisted extraction (HHPAE), supercritical fluid extraction (SFE), and extraction with eutectic solvents. In this body of work, Chanioti and Tzia [23] reported that the use of eutectic solvents improved the phenolic extraction yield from OP with respect to conventional solvents (i.e., solutions of water and ethanol). However, the content of the phenolics

that were obtained in the present work with real EOP that was extracted by ultrasound with an ethanol-water mixture of 43% (57.5 mg GAE/g EOP) was higher than that reported by these authors with the use of deep eutectic solvents and ultrasound as the extraction method (20.1 mg GAE/g OP). Caballero et al. (2018) [22] reported a TPC of 14.1 mg GAE from exhausted pomace while using SFE. Seçmeler et al. (2018) [28] proposed the use of steam explosion and other hydrothermal pretreatments, such as subcritical water for the recovery of phenols and reported higher values (69.7–73.3 mg GAE) than those that were obtained in the present research. Regarding the content of specific phenolic compounds, Xie et al., [25] found that UAE resulted in greater extraction e fficiencies of hydroxytyrosol, maslinic acid, and oleanolic acid from OP when compared with conventional extraction and MAE. Importantly, the comparison of the results from di fferent works that utilised di fferent extraction techniques is di fficult due to the variability of OP and EOP samples (e.g., di fferences due to the type of cultivation and the variety and maturation of the olives) and the di fferent methods that are used to quantify the phenolic compound content and measure the antioxidant activity. However, the results that were obtained here and in the studies reported in the bibliography sugges<sup>t</sup> that the residual biomass contains noticeable amounts of phenols and high antioxidant activity. Therefore, the extract from olive pomace has potential for use as a natural bioactive ingredient in di fferent industrial applications.
